On 07/17/15 12:10, Jasper Lievisse Adriaanse wrote:
> Hi,
>
> Here's a diff to add the '-i' flag to sed to do inplace edits. It's mostly
> from FreeBSD with some adjustments to prevent a race with unlink() and fopen()
> during the tempfile creation.
>
> It's been tested in a full ports bulk (thanks aja), and went through a build
> of base and xenocara.
> Regress tests will also be added for this.
>
> This diff is already OK millert@. Any more OKs?
>
Cool. This saves me having to install coreutils just to work on some
school projects.
ok bcallah@, fwiw.
> Index: defs.h
> ===================================================================
> RCS file: /cvs/src/usr.bin/sed/defs.h,v
> retrieving revision 1.5
> diff -u -p -r1.5 defs.h
> --- defs.h 19 Jan 2015 15:30:52 -0000 1.5
> +++ defs.h 16 Jul 2015 18:45:58 -0000
> @@ -128,6 +128,7 @@ typedef struct {
> char *space; /* Current space pointer. */
> size_t len; /* Current length. */
> int deleted; /* If deleted. */
> + int append_newline; /* If originally terminated by \n. */
> char *back; /* Backing memory. */
> size_t blen; /* Backing memory length. */
> } SPACE;
> Index: extern.h
> ===================================================================
> RCS file: /cvs/src/usr.bin/sed/extern.h,v
> retrieving revision 1.9
> diff -u -p -r1.9 extern.h
> --- extern.h 13 Apr 2015 05:11:23 -0000 1.9
> +++ extern.h 16 Jul 2015 00:23:57 -0000
> @@ -40,17 +40,19 @@ extern regmatch_t *match;
> extern size_t maxnsub;
> extern u_long linenum;
> extern size_t appendnum;
> -extern int lastline;
> extern int Eflag, aflag, eflag, nflag;
> -extern char *fname;
> +extern const char *fname, *outfname;
> +extern FILE *infile, *outfile;
>
> void cfclose(struct s_command *, struct s_command *);
> void compile(void);
> -void cspace(SPACE *, char *, size_t, enum e_spflag);
> +void cspace(SPACE *, const char *, size_t, enum e_spflag);
> char *cu_fgets(char **, size_t *);
> void err(int, const char *, ...);
> int mf_fgets(SPACE *, enum e_spflag);
> +int lastline(void);
> void process(void);
> +void resetranges(void);
> char *strregerror(int, regex_t *);
> void *xmalloc(size_t);
> void *xreallocarray(void *, size_t, size_t);
> Index: main.c
> ===================================================================
> RCS file: /cvs/src/usr.bin/sed/main.c,v
> retrieving revision 1.18
> diff -u -p -r1.18 main.c
> --- main.c 26 Nov 2014 18:34:51 -0000 1.18
> +++ main.c 16 Jul 2015 19:21:16 -0000
> @@ -34,6 +34,7 @@
> */
>
> #include <sys/types.h>
> +#include <sys/stat.h>
>
> #include <ctype.h>
> #include <errno.h>
> @@ -45,6 +46,7 @@
> #include <stdlib.h>
> #include <string.h>
> #include <unistd.h>
> +#include <libgen.h>
>
> #include "defs.h"
> #include "extern.h"
> @@ -78,15 +80,23 @@ struct s_flist {
> */
> static struct s_flist *files, **fl_nextp = &files;
>
> +FILE *infile; /* Current input file */
> +FILE *outfile; /* Current output file */
> +
> int Eflag, aflag, eflag, nflag;
> +static int rval; /* Exit status */
>
> /*
> * Current file and line number; line numbers restart across compilation
> - * units, but span across input files.
> + * units, but span across input files. The latter is optional if editing
> + * in place.
> */
> -char *fname; /* File name. */
> +const char *fname; /* File name. */
> +const char *outfname; /* Output file name */
> +static char oldfname[PATH_MAX]; /* Old file name (for in-place editing)
> */
> +static char tmpfname[PATH_MAX]; /* Temporary file name (for in-place
> editing) */
> +char *inplace; /* Inplace edit file extension */
> u_long linenum;
> -int lastline; /* TRUE on the last line of the last
> file */
>
> static void add_compunit(enum e_cut, char *);
> static void add_file(char *);
> @@ -97,7 +107,8 @@ main(int argc, char *argv[])
> int c, fflag;
>
> fflag = 0;
> - while ((c = getopt(argc, argv, "Eae:f:nru")) != -1)
> + inplace = NULL;
> + while ((c = getopt(argc, argv, "Eae:f:i::nru")) != -1)
> switch (c) {
> case 'E':
> case 'r':
> @@ -114,6 +125,9 @@ main(int argc, char *argv[])
> fflag = 1;
> add_compunit(CU_FILE, optarg);
> break;
> + case 'i':
> + inplace = optarg ? optarg : "";
> + break;
> case 'n':
> nflag = 1;
> break;
> @@ -123,8 +137,8 @@ main(int argc, char *argv[])
> default:
> case '?':
> (void)fprintf(stderr,
> - "usage: sed [-aEnru] command [file ...]\n"
> - " sed [-aEnru] [-e command] [-f command_file]
> [file ...]\n");
> + "usage: sed [-aEnru] [-i [extension]] command [file
> ...]\n"
> + " sed [-aEnru] [-i [extension]] [-e command]
> [-f command_file] [file ...]\n");
> exit(1);
> }
> argc -= optind;
> @@ -148,7 +162,7 @@ main(int argc, char *argv[])
> cfclose(prog, NULL);
> if (fclose(stdout))
> err(FATAL, "stdout: %s", strerror(errno));
> - exit (0);
> + exit (rval);
> }
>
> /*
> @@ -258,69 +272,128 @@ again:
> int
> mf_fgets(SPACE *sp, enum e_spflag spflag)
> {
> - static FILE *f; /* Current open file */
> + struct stat sb;
> size_t len;
> char *p;
> - int c;
> + int c, fd;
> + static int firstfile;
>
> - if (f == NULL)
> - /* Advance to first non-empty file */
> - for (;;) {
> - if (files == NULL) {
> - lastline = 1;
> - return (0);
> - }
> - if (files->fname == NULL) {
> - f = stdin;
> - fname = "stdin";
> - } else {
> - fname = files->fname;
> - if ((f = fopen(fname, "r")) == NULL)
> - err(FATAL, "%s: %s",
> - fname, strerror(errno));
> + if (infile == NULL) {
> + /* stdin? */
> + if (files->fname == NULL) {
> + if (inplace != NULL)
> + err(FATAL, "-i may not be used with stdin");
> + infile = stdin;
> + fname = "stdin";
> + outfile = stdout;
> + outfname = "stdout";
> + }
> +
> + firstfile = 1;
> + }
> +
> + for (;;) {
> + if (infile != NULL && (c = getc(infile)) != EOF) {
> + (void)ungetc(c, infile);
> + break;
> + }
> + /* If we are here then either eof or no files are open yet */
> + if (infile == stdin) {
> + sp->len = 0;
> + return (0);
> + }
> + if (infile != NULL) {
> + fclose(infile);
> + if (*oldfname != '\0') {
> + if (rename(fname, oldfname) != 0) {
> + err(WARNING, "rename()");
> + unlink(tmpfname);
> + exit(1);
> + }
> + *oldfname = '\0';
> }
> - if ((c = getc(f)) != EOF) {
> - (void)ungetc(c, f);
> - break;
> + if (*tmpfname != '\0') {
> + if (outfile != NULL && outfile != stdout)
> + fclose(outfile);
> + outfile = NULL;
> + rename(tmpfname, fname);
> + *tmpfname = '\0';
> }
> - (void)fclose(f);
> + outfname = NULL;
> + }
> + if (firstfile == 0)
> files = files->next;
> + else
> + firstfile = 0;
> + if (files == NULL) {
> + sp->len = 0;
> + return (0);
> + }
> + fname = files->fname;
> + if (inplace != NULL) {
> + if (lstat(fname, &sb) != 0)
> + err(1, "%s", fname);
> + if (!S_ISREG(sb.st_mode))
> + err(FATAL, "%s: %s %s", fname,
> + "in-place editing only",
> + "works for regular files");
> + if (*inplace != '\0') {
> + strlcpy(oldfname, fname,
> + sizeof(oldfname));
> + len = strlcat(oldfname, inplace,
> + sizeof(oldfname));
> + if (len > sizeof(oldfname))
> + err(FATAL, "%s: name too long", fname);
> + }
> + len = snprintf(tmpfname, sizeof(tmpfname),
> "%s/.%s.XXXXXXXXXX",
> + dirname(fname), basename(fname));
> + if (len >= sizeof(tmpfname))
> + err(FATAL, "%s: name too long", fname);
> + if ((fd = mkstemp(tmpfname)) == -1)
> + err(FATAL, "%s", fname);
> + if ((outfile = fdopen(fd, "w")) == NULL) {
> + unlink(tmpfname);
> + err(FATAL, "%s", fname);
> + }
> + fchown(fileno(outfile), sb.st_uid, sb.st_gid);
> + fchmod(fileno(outfile), sb.st_mode & ALLPERMS);
> + outfname = tmpfname;
> + linenum = 0;
> + resetranges();
> + } else {
> + outfile = stdout;
> + outfname = "stdout";
> + }
> + if ((infile = fopen(fname, "r")) == NULL) {
> + err(WARNING, "%s", fname);
> + rval = 1;
> + continue;
> }
> -
> - if (lastline) {
> - sp->len = 0;
> - return (0);
> }
>
> /*
> + * We are here only when infile is open and we still have something
> + * to read from it.
> + *
> * Use fgetln so that we can handle essentially infinite input data.
> * Can't use the pointer into the stdio buffer as the process space
> * because the ungetc() can cause it to move.
> */
> - p = fgetln(f, &len);
> - if (ferror(f))
> + p = fgetln(infile, &len);
> + if (ferror(infile))
> err(FATAL, "%s: %s", fname, strerror(errno ? errno : EIO));
> + if (len != 0 && p[len - 1] == '\n') {
> + sp->append_newline = 1;
> + len--;
> + } else if (!lastline()) {
> + sp->append_newline = 1;
> + } else {
> + sp->append_newline = 0;
> + }
> cspace(sp, p, len, spflag);
>
> linenum++;
> - /* Advance to next non-empty file */
> - while ((c = getc(f)) == EOF) {
> - (void)fclose(f);
> - files = files->next;
> - if (files == NULL) {
> - lastline = 1;
> - return (1);
> - }
> - if (files->fname == NULL) {
> - f = stdin;
> - fname = "stdin";
> - } else {
> - fname = files->fname;
> - if ((f = fopen(fname, "r")) == NULL)
> - err(FATAL, "%s: %s", fname, strerror(errno));
> - }
> - }
> - (void)ungetc(c, f);
> +
> return (1);
> }
>
> @@ -353,4 +426,52 @@ add_file(char *s)
> *fl_nextp = fp;
> fp->fname = s;
> fl_nextp = &fp->next;
> +}
> +
> +
> +static int
> +next_files_have_lines()
> +{
> + struct s_flist *file;
> + FILE *file_fd;
> + int ch;
> +
> + file = files;
> + while ((file = file->next) != NULL) {
> + if ((file_fd = fopen(file->fname, "r")) == NULL)
> + continue;
> +
> + if ((ch = getc(file_fd)) != EOF) {
> + /*
> + * This next file has content, therefore current
> + * file doesn't contains the last line.
> + */
> + ungetc(ch, file_fd);
> + fclose(file_fd);
> + return (1);
> + }
> +
> + fclose(file_fd);
> + }
> +
> + return (0);
> +}
> +
> +int
> +lastline(void)
> +{
> + int ch;
> +
> + if (feof(infile)) {
> + return !(
> + (inplace == NULL) &&
> + next_files_have_lines());
> + }
> + if ((ch = getc(infile)) == EOF) {
> + return !(
> + (inplace == NULL) &&
> + next_files_have_lines());
> + }
> + ungetc(ch, infile);
> + return (0);
> }
> Index: process.c
> ===================================================================
> RCS file: /cvs/src/usr.bin/sed/process.c,v
> retrieving revision 1.23
> diff -u -p -r1.23 process.c
> --- process.c 18 Apr 2015 18:28:37 -0000 1.23
> +++ process.c 16 Jul 2015 18:50:40 -0000
> @@ -55,6 +55,7 @@ static SPACE HS, PS, SS;
> #define pd PS.deleted
> #define ps PS.space
> #define psl PS.len
> +#define psanl PS.append_newline
> #define hs HS.space
> #define hsl HS.len
>
> @@ -76,7 +77,10 @@ static regex_t *defpreg;
> size_t maxnsub;
> regmatch_t *match;
>
> -#define OUT(s) do { fwrite(s, sizeof(u_char), psl, stdout); } while (0)
> +#define OUT() do {\
> + fwrite(ps, 1, psl, outfile);\
> + if (psanl) fputc('\n', outfile);\
> +} while (0)
>
> void
> process(void)
> @@ -85,6 +89,7 @@ process(void)
> SPACE tspace;
> size_t len, oldpsl;
> char *p;
> + int oldpsanl;
>
> for (linenum = 0; mf_fgets(&PS, REPLACE);) {
> pd = 0;
> @@ -118,8 +123,8 @@ redirect:
> case 'c':
> pd = 1;
> psl = 0;
> - if (cp->a2 == NULL || lastaddr)
> - (void)printf("%s", cp->t);
> + if (cp->a2 == NULL || lastaddr || lastline())
> + (void)fprintf(outfile, "%s", cp->t);
> break;
> case 'd':
> pd = 1;
> @@ -128,7 +133,7 @@ redirect:
> if (pd)
> goto new;
> if (psl == 0 ||
> - (p = memchr(ps, '\n', psl - 1)) == NULL) {
> + (p = memchr(ps, '\n', psl)) == NULL) {
> pd = 1;
> goto new;
> } else {
> @@ -140,25 +145,25 @@ redirect:
> cspace(&PS, hs, hsl, REPLACE);
> break;
> case 'G':
> - if (hs == NULL)
> - cspace(&HS, "\n", 1, REPLACE);
> + cspace(&PS, "\n", 1, 0);
> cspace(&PS, hs, hsl, 0);
> break;
> case 'h':
> cspace(&HS, ps, psl, REPLACE);
> break;
> case 'H':
> + cspace(&HS, "\n", 1, 0);
> cspace(&HS, ps, psl, 0);
> break;
> case 'i':
> - (void)printf("%s", cp->t);
> + (void)fprintf(outfile, "%s", cp->t);
> break;
> case 'l':
> lputs(ps);
> break;
> case 'n':
> if (!nflag && !pd)
> - OUT(ps);
> + OUT();
> flush_appends();
> if (!mf_fgets(&PS, REPLACE))
> exit(0);
> @@ -166,33 +171,32 @@ redirect:
> break;
> case 'N':
> flush_appends();
> - if (!mf_fgets(&PS, 0)) {
> - if (!nflag && !pd)
> - OUT(ps);
> + cspace(&PS, "\n", 1, 0);
> + if (!mf_fgets(&PS, 0))
> exit(0);
> - }
> break;
> case 'p':
> if (pd)
> break;
> - OUT(ps);
> + OUT();
> break;
> case 'P':
> if (pd)
> break;
> - if (psl != 0 &&
> - (p = memchr(ps, '\n', psl - 1)) != NULL) {
> + if ((p = memchr(ps, '\n', psl)) != NULL) {
> oldpsl = psl;
> - psl = (p + 1) - ps;
> - OUT(ps);
> + oldpsanl = psanl;
> + psl = p - ps;
> + psanl = 1;
> + OUT();
> psl = oldpsl;
> } else {
> - OUT(ps);
> + OUT();
> }
> break;
> case 'q':
> if (!nflag && !pd)
> - OUT(ps);
> + OUT();
> flush_appends();
> exit(0);
> case 'r':
> @@ -225,34 +229,36 @@ redirect:
> DEFFILEMODE)) == -1)
> err(FATAL, "%s: %s",
> cp->t, strerror(errno));
> - if (write(cp->u.fd, ps, psl) != psl)
> + if (write(cp->u.fd, ps, psl) != psl ||
> + write(cp->u.fd, "\n", 1) != 1)
> err(FATAL, "%s: %s",
> cp->t, strerror(errno));
> break;
> case 'x':
> if (hs == NULL)
> - cspace(&HS, "\n", 1, REPLACE);
> + cspace(&HS, "", 0, REPLACE);
> tspace = PS;
> PS = HS;
> + psanl = tspace.append_newline;
> HS = tspace;
> break;
> case 'y':
> if (pd || psl == 0)
> break;
> - for (p = ps, len = psl; --len; ++p)
> + for (p = ps, len = psl; len--; ++p)
> *p = cp->u.y[(unsigned char)*p];
> break;
> case ':':
> case '}':
> break;
> case '=':
> - (void)printf("%lu\n", linenum);
> + (void)fprintf(outfile, "%lu\n", linenum);
> }
> cp = cp->next;
> } /* for all cp */
>
> new: if (!nflag && !pd)
> - OUT(ps);
> + OUT();
> flush_appends();
> } /* for all lines */
> }
> @@ -263,7 +269,7 @@ new: if (!nflag && !pd)
> */
> #define MATCH(a) \
> (a)->type == AT_RE ? regexec_e((a)->u.r, ps, 0, 1, psl) : \
> - (a)->type == AT_LINE ? linenum == (a)->u.l : lastline
> + (a)->type == AT_LINE ? linenum == (a)->u.l : lastline()
>
> /*
> * Return TRUE if the command applies to the current line. Sets the inrange
> @@ -305,6 +311,19 @@ applies(struct s_command *cp)
> }
>
> /*
> + * Reset all inrange markers.
> + */
> +void
> +resetranges(void)
> +{
> + struct s_command *cp;
> +
> + for (cp = prog; cp; cp = cp->code == '{' ? cp->u.c : cp->next)
> + if (cp->a2)
> + cp->inrange = 0;
> +}
> +
> +/*
> * substitute --
> * Do substitutions in the pattern space. Currently, we build a
> * copy of the new pattern space in the substitute space structure
> @@ -392,19 +411,21 @@ substitute(struct s_command *cp)
> */
> tspace = PS;
> PS = SS;
> + psanl = tspace.append_newline;
> SS = tspace;
> SS.space = SS.back;
>
> /* Handle the 'p' flag. */
> if (cp->u.s->p)
> - OUT(ps);
> + OUT();
>
> /* Handle the 'w' flag. */
> if (cp->u.s->wfile && !pd) {
> if (cp->u.s->wfd == -1 && (cp->u.s->wfd = open(cp->u.s->wfile,
> O_WRONLY|O_APPEND|O_CREAT|O_TRUNC, DEFFILEMODE)) == -1)
> err(FATAL, "%s: %s", cp->u.s->wfile, strerror(errno));
> - if (write(cp->u.s->wfd, ps, psl) != psl)
> + if (write(cp->u.s->wfd, ps, psl) != psl ||
> + write(cp->u.s->wfd, "\n", 1) != 1)
> err(FATAL, "%s: %s", cp->u.s->wfile, strerror(errno));
> }
> return (1);
> @@ -425,7 +446,7 @@ flush_appends(void)
> switch (appends[i].type) {
> case AP_STRING:
> fwrite(appends[i].s, sizeof(char), appends[i].len,
> - stdout);
> + outfile);
> break;
> case AP_FILE:
> /*
> @@ -439,12 +460,12 @@ flush_appends(void)
> if ((f = fopen(appends[i].s, "r")) == NULL)
> break;
> while ((count = fread(buf, sizeof(char), sizeof(buf),
> f)))
> - (void)fwrite(buf, sizeof(char), count, stdout);
> + (void)fwrite(buf, sizeof(char), count, outfile);
> (void)fclose(f);
> break;
> }
> - if (ferror(stdout))
> - err(FATAL, "stdout: %s", strerror(errno ? errno : EIO));
> + if (ferror(outfile))
> + err(FATAL, "%s: %s", outfname, strerror(errno ? errno : EIO));
> appendx = sdone = 0;
> }
>
> @@ -452,10 +473,14 @@ static void
> lputs(char *s)
> {
> int count;
> - char *escapes, *p;
> + const char *escapes;
> + char *p;
> struct winsize win;
> static int termwidth = -1;
>
> + if (outfile != stdout)
> + termwidth = 60;
> +
> if (termwidth == -1) {
> termwidth = 0;
> if ((p = getenv("COLUMNS")))
> @@ -470,29 +495,33 @@ lputs(char *s)
>
> for (count = 0; *s; ++s) {
> if (count >= termwidth) {
> - (void)printf("\\\n");
> + (void)fprintf(outfile, "\\\n");
> count = 0;
> }
> if (isascii((unsigned char)*s) && isprint((unsigned char)*s)
> && *s != '\\') {
> - (void)putchar(*s);
> + (void)fputc(*s, outfile);
> count++;
> - } else if (*s != '\n') {
> + } else if (*s == '\n') {
> + (void)fputc('$', outfile);
> + (void)fputc('\n', outfile);
> + count = 0;
> + } else {
> escapes = "\\\a\b\f\r\t\v";
> - (void)putchar('\\');
> + (void)fputc('\\', outfile);
> if ((p = strchr(escapes, *s))) {
> - (void)putchar("\\abfrtv"[p - escapes]);
> + (void)fputc("\\abfrtv"[p - escapes], outfile);
> count += 2;
> } else {
> - (void)printf("%03o", *(u_char *)s);
> + (void)fprintf(outfile, "%03o", *(u_char *)s);
> count += 4;
> }
> }
> }
> - (void)putchar('$');
> - (void)putchar('\n');
> - if (ferror(stdout))
> - err(FATAL, "stdout: %s", strerror(errno ? errno : EIO));
> + (void)fputc('$', outfile);
> + (void)fputc('\n', outfile);
> + if (ferror(outfile))
> + err(FATAL, "%s: %s", outfname, strerror(errno ? errno : EIO));
> }
>
> static inline int
> @@ -507,9 +536,7 @@ regexec_e(regex_t *preg, const char *str
> } else
> defpreg = preg;
>
> - /* Set anchors, discounting trailing newline (if any). */
> - if (slen > 0 && string[slen - 1] == '\n')
> - slen--;
> + /* Set anchors */
> match[0].rm_so = 0;
> match[0].rm_eo = slen;
>
> @@ -575,7 +602,7 @@ regsub(SPACE *sp, char *string, char *sr
> * space as necessary.
> */
> void
> -cspace(SPACE *sp, char *p, size_t len, enum e_spflag spflag)
> +cspace(SPACE *sp, const char *p, size_t len, enum e_spflag spflag)
> {
> size_t tlen;
>
> Index: sed.1
> ===================================================================
> RCS file: /cvs/src/usr.bin/sed/sed.1,v
> retrieving revision 1.44
> diff -u -p -r1.44 sed.1
> --- sed.1 22 Oct 2014 23:23:22 -0000 1.44
> +++ sed.1 16 Jul 2015 19:15:14 -0000
> @@ -47,6 +47,7 @@
> .Op Fl aEnru
> .Op Fl e Ar command
> .Op Fl f Ar command_file
> +.Op Fl i Op Ar extension
> .Op Ar
> .Sh DESCRIPTION
> The
> @@ -94,6 +95,16 @@ Append the editing commands found in the
> .Ar command_file
> to the list of commands.
> The editing commands should each be listed on a separate line.
> +.It Fl i Ar extension
> +Edit files in-place, saving backups with the specified
> +.Ar extension .
> +If a zero-length
> +.Ar extension
> +is given, no backup will be saved.
> +It is not recommended to give a zero-length
> +.Ar extension
> +when in-place editing files, as you risk corruption or partial content
> +in situations where disk space is exhausted, etc.
> .It Fl r
> An alias for
> .Fl E ,
> @@ -510,6 +521,12 @@ command,
> squeezing excess empty lines from standard input:
> .Bd -literal -offset indent
> $ sed -n '
> +.Pp
> +The
> +.Fl i
> +option is a non-standard
> +.Fx
> +extension and may not be available on other operating systems.
> # Write non-empty lines.
> /./ {
> p
> @@ -543,7 +560,7 @@ utility is compliant with the
> specification.
> .Pp
> The flags
> -.Op Fl aEru
> +.Op Fl aEiru
> are extensions to that specification.
> .Pp
> The use of newlines to separate multiple commands on the command line
>
